Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

Bronze

Reg OSPF behavior

In OSPF Hello process, if any of below mentioned item does not match, two routers do not form neighbor relationship:-

- Must Pass Authentication

- Must be in the same primary subnet, including subnet mask

- Must be in same OSPF Area

- Must be of same area type

- Must not have duplicate RIDs

- OSPF Hello and Dead timers must be equal

As all the mentioned parameters are exchanged and checked in Hello packets, if not matched, router will not become neighbors

Now the test scenario

1) Take a case of frame relay network

If we define interface type at one end as broadcast and other as point to point, as the default timers for both the interface type are same so neighbor relation should come up. But do they become adjacent.

I think they will become adjacent as interface type check is not done anytime during the OSPF adjacency process, just the related parameter i.e. timers are checked (that we made sure are equal)

neighbor status (show ip ospf neighbor) will be "Full/-" at P2P end and "Full/DROther? at broadcast defined interface. Broadcast will think its neighbor a DROther.

First question: - Is the above explanation correct?

Second question: - What are the operational issues with interface type mismatch even if both the router are adjacent now and why there are issues?

2) Similarly if we check the same scenario in an Ethernet network i.e. interface mismatch case with same default timer? Will the behavior be different from frame relay?

1 ACCEPTED SOLUTION

Accepted Solutions
Cisco Employee

Re: Reg OSPF behavior

Rajat,

The two routers have exactly identical LSDB. It is just one router LSA sees the link between them has p2p and the other has broadcast. The SPF calculation on each router will therefore not use this perticular link.

Hope this helps,

Harold Ritter
Sr. Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México 
Paseo de la Reforma 222 Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México
9 REPLIES

Re: Reg OSPF behavior

Hi,

The network type shouldn't be the same but only should be compatible, meaning that if one of them supports DR/BDR then the other must do, and accordingly a broadcast interface is compatible with a NBMA interface, and similarly point-to-point, point-to-multipoint and point-to-multipoint non-broadcast are also compatible - but always take care of the timers as they must match and they differ by default for different network types.

Also please note that practically speaking incompatible network types gets the adjacency up (with repect to each network type) and LSAs are exchanged but never entered into the routing table of the neighbor.

HTH,

Mohammed Mahmoud.

Bronze

Re: Reg OSPF behavior

Hi,

If I understood it correctly :-

you are saying that if all the mentioned parameters match including timers, so even if the interface types are different (we have taken a case of broadcast at one end and P2P at other end) , adjacency will come up.

So the first thing is that my understanding of this scenario is correct..great

Now lets move forward...

Also you said LSAs are exchanged but never entered into the routing table (so thats the answer to my second question about any operational issues) but WHY donnt they enter the routing table ?

If the full adjacency has come up that means both neighbors LSA database match and so when the interface type mismatch comes into picture ?

Apologies but I donnt have any lab to test these scenarios so i am troubling the experts here rather than digging into debugging logs :) .... hope you understand

rgds

Re: Reg OSPF behavior

Hi,

You are very welcomed :) and yes you've got it all right, perhaps the following document can clarify more, this kind of network type mismatch makes the advertising router unreachable.

http://www.cisco.com/warp/public/104/26.html#r1

I hope that i've been informative.

HTH,

Mohammed Mahmoud.

Bronze

Re: Reg OSPF behavior

Hi Mohammed,

I was looking forward for this kind of discussion and your information helped a lot.

and the doc you referred has this statement

"You can see that for subnet 172.16.32.0/24, Router R1-7010 is generating a point-to-point link and Router R4-4K is generating a transit link. This creates a discrepancy in the link-state database, which means no routes are installed in the routing table"

So that means discrepancy in LSDB doesnt allow the entry to routing table.

But , sorry to say there is again a question mark on this.

If there is discrepancy in LSDBs of both router they should not become adjacent.

As any two router will be adjacent if they have identicle LSDB....???

Let me know your inputs on this ...

rgds

Cisco Employee

Re: Reg OSPF behavior

Rajat,

The two routers have exactly identical LSDB. It is just one router LSA sees the link between them has p2p and the other has broadcast. The SPF calculation on each router will therefore not use this perticular link.

Hope this helps,

Harold Ritter
Sr. Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México 
Paseo de la Reforma 222 Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México

Re: Reg OSPF behavior

Hi Rajat,

I think that Harold has answered your question clearly, the LSDB is identical but the routers will choose not to install the routes into the routing table, this is what is meant by "discrepancy in the link-state database".

HTH,

Mohammed Mahmoud.

Re: Reg OSPF behavior

Hi Rajat,

Sorry for the rush i wrote my last post in, you are very welcomed :) and i hope that its clear for you now.

HTH,

Mohammed Mahmoud.

Bronze

Re: Reg OSPF behavior

Thanks Harold and Mohammed!!

I got it now, router will use SPF algorithim to examine the LSAs in its database and will derive its own topology of network.

Now spf calculation will result in discrepancy in LSDB.

e.g.

After SPF calculation based on LSAs, P2P interface will see its adjacent neighbor defined a Trasient Link, This will bring up the conflict.

Dear Mohammed,

No need to say sorry, It was good dicussion and you guys really helped me in understanding the root concept.

Thx and Rgds

Re: Reg OSPF behavior

Dear Rajat,

You've got it right, and you are very welcomed any time.

HTH,

Mohammed Mahmoud.

178
Views
9
Helpful
9
Replies